eric.base.eth
@0xEricBrown
agents + devrel @base + @CoinbaseDev
≈≈≈
参加 February 2019
1.7K フォロー中 4.2K ファン
A lot of praise for this bench and I agree
most representative of experience daily driving different models
Opus 4.8 is now on DeepSWE.
On the default high thinking effort, it scores 6% higher than Opus 4.7 xhigh, while also lowering average cost per task.
